A comparison to schemes of ocean surface albedo parameterization and their impact on shortwave radiatation estimation

Abstract:

The ocean covers 71% of the Earth’s surface and plays a pivotal role in the earth radiation energy balance. The ocean surface albedo(OSA) is a deciding factor on ocean net surface shortwave radiation(ONSSR) estimation. Several OSA schemes have been proposed successively, but there is not a conclusion for the best OSA scheme of estimating the ONSSR. This study, on the base of analyzing currently existing OSA parameterization, including Briegleb et al.(B), Taylor et al.(T), Hansen et al.(H), Jin et al.(J), Preisendorfer and Mobley(PM86), Feng’s scheme(F), discusses the difference of OSA’s impact on ONSSR estimation in condition of actual downward shortwave radiation(DSR). Then we evaluate the necessity and applicability for the climate models to integrate the more complicated OSA scheme.

We got some conclusions: The SZA and the wind speed are the two most significant effect factor to broadband OSA, thus the different OSA parameterizations varies violently in the regions of both high latitudes and strong winds. In the summer, the Northern Hemisphere(NH) is high ONSSR, but small deviations compared with Northern Hemisphere(SH),and contrary in the winter. The OSA schemes can lead the ONSSR results difference of the order of 20 w m^-2 by the analysis of the ONSSR reanalysis dataset, the Modern Era Retrospective-analysis for Research and Applications (MERRA).The simple scheme of Taylor and the more complicate schemes of Jin and Feng is very similar, and the scheme B and H is close to each other, the PM86 is more close to MERRA. We use the COVE ocean platform observation data to validate the several scheme result, and the RMSE is 10.96 w m^-2, 5.24 w m^-2, 12.88 w m^-2, 6.52 w m^-2, 6.33 w m^-2, 6.30 w m^-2 for B,T,H,J,PM86,F, respectively. The Taylor’s scheme shows the best estimate, and Feng’s result just following Taylor’s. However, the accuracy of the estimated instantaneous OSA changes at different local time. Jin’s scheme has the best performance generally at noon and in the afternoon, and PM86’s is the best of all in the morning, which indicate that the more complicated OSA schemes reflect the temporal variation of OWA better than the simple ones.